Design of more potent and selective antagonists of the antidiuretic

structural features at position 2 required for antidiuretic antagonism, we have synthesizedeight new analogues substituted at position 2 by the solid-...
0 downloads 0 Views 876KB Size
414

J.Med. Chem. 1982,25,414-419

Amino acid analysis: Tyr, 0.98; Phe, 0.98; Val, 1.03; Asp, 1.01; Pro, 1.01; Arg, 1.01; Gly, 1.00; NH8, 2.11. Analysis following performic acid oxidation prior hydrolysis gave a Cys(O,H)-Gly ratio of 1.05:l.OO.

Acknowledgment. This work was supported in part by research grants from the National Institutes of General

Medical Sciences (GM-25280), the National Institute of Arthritis, Metabolism and Digestive Diseases (AM-01940), and t h e National H e a r t a n d Lung Institute (HL-12738). The authors thank Dr T. C. Wuu for generous use of amino acid analysis facilities and Ms. Beverly Lockwood for assistance in t h e preparation of t h e manuscript.

Design of More Potent and Selective Antagonists of the Antidiuretic Responses to Arginine-vasopressin Devoid of Antidiuretic Agonism Maurice Manning,***Wieslaw A. Klis,*J Aleksandra Olma,*~llJ a n n y Seta,§ a n d Wilbur H. Sawyers Department of Biochemistry, Medical College of Ohio at Toledo, Toledo, Ohio 43699, and Department of Pharmacology, College of Physicians and Surgeons of Columbia University, New York, New York 10032. Received October 23, 1981

Substitution of D-tyrosine at position 2 of [ 1-(0-rnercapto-P,P-cyclopentamethylenepropionic acid),4-valine]arginine-vasopressin,d(CH2),VAVP,turned this weak antidiuretic agonist and antagonist of vasopressor responses into an effective antagonist of the antidiuretic response. d(CH2),-~-TyrVAVP, however, like other reported antagonists of the antidiuretic response, retains some antidiuretic agonistic activity. It is also a relatively strong antagonist of vasopressor and oxytocic responses. In attempting (a) to increase the specificity of antagonists of the antidiuretic response, (b) to eliminate residual agonistic activity and enhance antagonistic potency, and (c) to help delineate structural features at position 2 required for antidiuretic antagonism, we have synthesized eight new analogues substituted at position 2 by the solid-phase method 1, d(CH2),-~-PheVAVP;2, d(CH2),-D-PheVDAVP; 3, d(CH~)~-D-I~~VAVP; 4, d(CH2),-~-LeuVAVP; 5, d(cH&-~-vah’AVP; 6, d(CH2),-D-AlaVAVP;7, d(CH,),GlyVAVP; 8, d(CH&D-ArgVAVP. These analogues were tested for agonistic and antagonistic activities by antidiuretic, vasopressor, and oxytocic assays in rats. Analogues 1,3,4, and 5 exhibit no agonistic activities in these assays. These four analogues, as well as analogue 2, effectively antagonize antidiuretic responses to AVP. Their antiantidiuretic pA2 values are as follows: 1, 8.07 f 0.09; 2, 7.07 0.1; 3, 7.98 h 0.05;4,7.79 h 0.12; 5,7.48 h 0.06. Analogues 6-8 are weak antidiuretic agonists and exhibit no detectable antiantidiuretic activity. Analogues 3-8 show greatly reduced potencies as antagonists of vasopressor and oxytocic responses. Thus, analogues 3-5 show much greater specificity as antagonists of the antidiuretic response than any previously reported. Analogues 1 and 3 are also the most potent antagonists of the antidiuretic response yet reported. The combination of increased antiantidiuretic potency and specificity should make these analogues useful tools for studies on the role of AVP in causing water retention in experimental animals and in man. They may also serve as prototypes for the design of even more potent and selective antidiuretic antagonists. Potent and specific antagonists of the antidiuretic action of ADH could be valuable for treating water retention in a variety of clinical situations.

*

We recently reported t h e first known effective antagonists of in vivo antidiuretic responses to both exogenous a n d endogenous arginine-vasopressin (AVP) . l i 2 These antagonists were designed by incorporating 0-alkyltyrosine residues (where alkyl = methyl, ethyl, isopropyl, or npropyl) at position 2 in two vasopressor antagonists, [ l (P-mercapto-P,P-cyclopentamethylenepropionic acid),4valine,8-D-arginine]vasopressin[d(CH2),VDAVPI3 and [ 1-(P-mercapto-P,P-cyclopentamethylenepropionic acid),4-valine]arginine-vasopressin[d(CH,),VAVP]? We subsequently found that the substitution of a n O-alkylD-tyrosine residue in each of these eight antidiuretic antagonists brought about substantial increases in their respective antiantidiuretic p ~ t e n c i e s . ~ *This ~ led t o the discovery that t h e incorporation of an unalkylated D-tyrosine residue for L-tyrosine at position 2 in d(CH2),VAVP and in d(CH,),VDAVP converted these weak antidiuretic agonists into potent antagonists of the antidiuretic responses t o AVP.,s6 All of these antidiuretic antagonists exhibit transient antidiuretic agonistic activity and are also potent antagonists of the vasopressor responses to AVP.295@ The finding that t h e substitution of a n unalkylated Dtyrosine residue in d(CHZ),VAVP and in d(CH,),VDAVP t Medical College of Ohio at Toledo. 1Visiting investigator from the University of Wroclaw, Poland. 11 Visiting investigator from the Technical University of Lodz, Poland. College of Physicians and Surgeons of Columbia University. 0022-2623/82/1825-0414$01.25/0

led to peptides possessing antidiuretic antagonism raised the possibility that t h e incorporation of other D-amino acids at position 2 i n d(CH,),VAVP a n d i n d(CH,),VDAVP might lead t o more potent and/or more selective antidiuretic antagonists lacking any residual antidiuretic agonistic activity. Initially, we substituted Dphenylalanine at position 2 in d(CH2),VAVP a n d in d(CH,),VDAVP. Preliminary data on the resulting peptides were highly encouraging; both peptides were effective antidiuretic antagonists. T h e L-arginine-containing peptide was, however, much more potent t h a n the D-argininecontaining one-a result highly consistent with all of our Thus, we decided t o restrict further exearlier data.1*2p6*6 plorations of position 2 with D-amino acid substitutions to d(CH,),VAVP. Therefore, following these findings with (1) Sawyer, W. H.; Pang, P. K. T.; Seto, J.; McEnroe, M.; Lammek, B.; Manning, M. Science 1981,212, 49. (2) Manning, M.; Lammek, B.; Kolodziejczyk, A. M.; Seto, J.;

Sawyer, W. H. J. Med. Chem. 1981,24,701. (3) Lowbridge, J.; Manning, M.; Haldar, J.; Sawyer, W. H. J. Med. Chem. 1978,21, 313. (4) Manning, M.; Lammek, B.; Kruszynski, M.; Seto, J.; Sawyer, W. H. J. Med. Chem., preceding paper in this issue. ( 5 ) Manning, M.; Olma, A,; Klis, W. A,; Kolodziejczyk,A. M.; Seto, J.; Sawyer, W. H. Proceedings of the American Peptide Symposium, 7th, Rich, D.; Gross, E., Eds.; Pierce Chemical Co.: Rockford, IL, 1981, p 257. (6) Manning, M.; Olma, A.; Klis, W. A.; Kolodziejczyk,A. M.; Seto, J.; Sawyer, W. H. J . Med. Chem. 1982, 25, 45.

0 1982 American Chemical Society

Journal of Medicinal Chemistry, 1982, Vol. 25, No. 4 415

Arginine-vasopressin Antagonists

modification of the standard workup procedure.2 The the D-Phe2-substituted analogues, we selected additional yields of free peptides were 2-3 times greater than those D-amino acids which would provide some insight into the of the 0-alkyltyrosine analogues.2,6 The deblocked distructural features of the side chains at position 2 required sulfhydryl compounds were oxidatively cyclized with K3for antagonism. We wondered, for example, if aromatic [Fe(CN),].le The analogues were desalted and purified character was a requirement. If not, how large did the by gel filtration on Sephadex G-15 as previously deside-chain have to be and could it contain an ionized scribed.17 grouping? The additional amino acids chosen to answer Bioassay Methods. The agonistic and antagonistic these questions were D-isoleucine,D-leucine, D-valine, Dpotencies of these analogues were measured using previalanine, and D-arginine. To test the possibility that a side ously described methods.1*2Jgm These included intravechain was not a requirement for antidiuretic antagonism, nous vasopressor assays in phenoxybenzamine-treated rats we decided to incorporate glycine at position 2. The eight under urethane anesthesia, antidiuretic assays in rats unnew analogues designed following the above reasoning are as follows: 1, [1-(@-mercapto-@,@-cyclopentamethylene- der ethanol anesthesia, and oxytocic assays on isolated rat propionic acid),2-~-phenylalanine,4-valine]arg~n~ne-vaso-uteri and rat uteri in situ. The USP posterior pituitary pressin [d(CH2),-~-PheVAVP];2, [1-(0-mercapto-@,@- reference standard was used in assays for agonistic and cyclopentamethylenepropionic acid),2-~-phenylalanine,4- antagonistic activities, except for the rat uterus in situ. Synthetic oxytocin (Syntocinon, Sandoz) was used as the valine,8-D-arginine]vasopressin[d(CH2),-~-PheVDAVP]; agonist in these assays. Agonistic activities are expressed 3, [1-(8-mercapto-@,@-cyclopentamethylenepropionic in units per milligram. Antagonistic potencies were deacid),2-~-isoleuc~ne,4-val~ne]argin~ne-vasopress~n [dtermined and expressed as “effective doses” and as pA2 (CH2)5-~-IleVAVP]; 4, [1-(@-mercapto-o,@-cyclopentamethylenepropionic acid),2-~-leucine,4-~ahe]arginine- values.21 The effective dose is defined as the dose (in vasopressin [d(CHZ),-~-LeuVAVP]; 5, [ 1-(@-mercapto-@,- nanomoles per kilogram) that reduces the response seen @-cyclopentamethylenepropionic acid),2-~-valine,4-va- from 2x units of agonist to the response with lx units of linelarginine-vasopressin [d(CH2),-~-ValVAVP]; 6, [1-(@- agonist. Estimated in vivo “pA2” values represent the negative logarithms of the effective doses divided by the mercapto-@,@-cyclopentamethylenepropionicacid),2-Dalanine,4-valine]arginine-vasopressin[d(CH2)5-~-AIa- estimated volume of distribution (67 mL/kg). Inhibition VAVP]; 7, [1-(0-mercapto-@,@-cyclopentamethylene- of antidiuretic responses was tested by injecting the standard 20 min after injecting the antagonist, to allow for propionic acid),2-glycine,4-valine]arginine-vasopressin [drecovery from the initial antidiuretic responses to some (CH,),GlyVAVP]; 8, [1-(@-mercapto-@,@-cyclopentaof the antagonists. For those analogues showing no methylenepropionic acid),2-~-arginine,4-valine] arginineagonistic activity (compounds 1,3,4,and 5) the standard vasopressin [d(CH2),-~-ArgVAVP].These eight analogues could be injected 10 min after the antagonist. Each anhave the following general structure: tagonist was administered in two doses, a high dose which 1 2 3 4 5 6 7 8 9 reduced the response to 2x units of agonist to less than the CH2-CO-(X)- P h e - V a l - A m - C y - Pro - A r g - G I y - N H z response to l x units of agonist, and a low dose which did not fully reduce the response to that given by l x units of agonist. The effective dose in each case was obtained by interpolation on a logarithmic scale between the two doses S no. 1 2* 3 4 5 6 7 8 of antag0nist.l

I

X = D-Phe, D-Phe, D-Ile, D-Leu, D-Val, D-Ala, Gly, D-Arg

Results and Discussion The antiantidiuretic, antivasopressor, and antioxytocic potencies of analogues 1-8 are presented as effective doses We now present the synthesis and some pharmacological and as pA2values in Tables I and 111. The effects of some properties of these eight analogues. of the D-amino acid substitutions at position 2 on antianPeptide Synthesis. The protected peptide precursors tidiuretic/antivasopressor and antiantidiuretic/antioxyrequired for the synthesis of all eight peptide antagonists tocic selectivities are presented in Table I1 and IV. Four were prepared by the solid-phase method of peptide synthesis.’-1° p-Nitrophenyl @-(S-benzy1mercapto)-@,@- of these analogues exhibit no detectable antidiuretic agonistic activity. These are d(CH2),-~-PheVAVP,dcyclopentamethylenepropionatellwas used in each final (CH2)5-~-IleVAVP, d(CH,),-D-LeuVAVP, and d(cH,),-~were facilitated coupling step. AII active ester c0up1ings’~J~ ValVAVP. These four analogues and d(CH2),-~-Phe by the addition of N-hydroxybenzotriazole(HOBT).14 All VDAVP reversibly antagonize subsequent doses of AVP of the protected peptide amides were obtained by amfor 1-3 h depending on the dose of antagonist. Two of monolytic cleavagegfrom the respective acyl octapeptide these peptides, d(CHJ5-~-PheVAVPand d(CH2),-~-Ileresins. Each protected precursor was deblocked with Na VAVP are more potent than any previously reported from in NH315 using a previously described yield-enhancing these l a b ~ r a t o r i e s . ’ ~All, ~ ~except ~ * ~ the Gly2 peptide, are antagonists of the vasopressor responses to AVP. However,

* D-&g

at position 8

~~~~

(7) (8) (9) (10) (11)

(12) (13) (14) (15)

Merrifield, R. B. J. Am. Chem. SOC.1963, 85, 2149. Merrifield, R. B. Biochemistry 1964, 3, 1385. Manning, M. J. Am. Chem. SOC.1968,90, 1348. Manning, M.; Coy, E.; Sawyer, W. H. Biochemistry 1970, 9, 3925. Nestor, Jr., J. J.; Ferger, M. F.; du Vigneaud, V. J. Med. Chem. 1975, 18, 284. Bodanszky, M.; du Vigneaud, V. J. Am. Chem. SOC.1959,81, 5688. Bodanszky, M.; Kondo, M.; Lin, C. Y.; Sigler, G. F. J . Org. Chem. 1974,39,44. Konig, W.; Geiger, R. Chem. Ber. 1970, 103, 788. du Vigneaud, V.; Ressler, C.; Swan, J. M.; Katsoyannis, P. G.; Roberts, C. W. J. Am. Chem. SOC.1954, 76, 3115.

-

(16) Hope, D. B.; Murti, V. V. 8.: du Vimeaud. V. J. Biol. Chem. 1962, 237, 1563. (17) Manning, M.; Wuu, T. C.; Baxter, J. W. M. J. Chromatogr. 1968, 38, 396. (18) Manning, M.; Lowbridge, L.; Stier, Jr., C. T.; Haldar, J.; Sawyer, W. H. J. Med. Chem. 1977,20, 1228. (19) Kruszynski, M.; Lammek, B.; Manning, M.; Seto, J.; Haldar, J.; Sawyer, W. H. J . Med. Chem. 1980, 23, 364. (20) Sawyer, W. H.; Haldar, J.; Gazis, D.; Seto, J.; Bankowski, K.; Lowbridge, J.; Turan, A.; Manning, M. Endocrinology 1980, 106, 81. (21) Schild, H. 0. Br. J . Pharmacol. Chemother. 1947,2, 189.

416 Journal of Medicinal Chemistry, 1982, Vol. 25, No. 4

Manning et al.

Table I. Antiantidiuretic and Antivasopressor Potencies of Arginine-vasopressin ( AVP) Antagonists Containing Some D-Amino Acids and Glycine at Position 2 ~~

I

2

3

4

5

6

7

8

.

9

CHz-CO-c -Phe -Phe-Val - A m - C y - Pro-Arg-Gly-NH2 CH2 -C H

C‘*

C HZ \cH,-cH~/

I

I s

S

d(CH,),-D-Phe VAVP antiantidiuretic antivasopressor effective dose,= effective dose,‘“ no. peptideg nmol/kg PA2 nmol/kg PA, d(CH,),VAVPd agonist 0.76 f 0.10 7.97 t 0.06 ( 8 ) c d(C H,),-D-TyrV AVPe 2.2 f 0.2 7.51 i. 0.08(4)c 0.29 f 0.09 8.41 * 0.11 (4) 6.3 c 0.8 7.03 i 0.05 (4) 0.60 i 0.04 8.05 0.03 (4) d(CH,),-D-TyrVDAVPe 1 d(CH,),-D-PheVAVP 0.67 i 0.13f 8.07 i 0.09 ( 8 ) 0.58 i. .04 8.06 i. 0.03 (4) 2 d(CH,),-D-PheVDAVP 6.9 t 1.3 7.07 i 0.10 ( 9 ) 0.73 7.98 i 0.07 (4) 3 d(CH,),-D-IleVAVP 0.70 i 0.01 7.98 i. 0.05 (4) 8.2 f 1.4 6.94 i 0.08 ( 5 ) 4 d( CH,),-D-LeuVAVP 1.2 ?r 0.3f 7.79 i 0.12 (4) 26i. 5 6.45 i 0.09 (4) 5 d(CH,),-D-ValVAVP 2.3 i 0.3f 7.48 rt 0.06 (4) 27 i 3 6.41 i 0.05 (4) 6 d(CH,),-D-AlaVAVP agonist 177 i. 31 5.79 1 0.08 ( 4 ) 7 d(CH,),GlyVAVP agonist agonist 8 d( C H,),-D -ArgV AVP >90